Corresponding author: Employing a recently developed method-mixed data sampling (MIDAS)
approach - to assess the risk-return trade-off for Chinese stock markets, our results
are striking. First, we fail to find any evidence of the risk-return trade-off in
the first subsample (Jan 1993–Jan 2001), while we do find the existence of such
relationship in the second subsample (Feb 2001–Dec 2005); such results suggest
that as the markets become more mature, risks are compensated more properly.
Second, we also compare the MIDAS results with the results obtained from
conventional approaches such as the GARCH-type model. Our results are reasonably
robust to the methods that we use, and the MIDAS and GARCH-type
approaches outperform rolling-window approach in terms of modeling volatility.
Show less